Dear authors,

there appears to be a misunderstanding of the process of peer-review for Biogeoscience Discussions. You were not meant to provide a revised manuscript at this point, and not meant to make this public as an author comment. Rather, the process foresees that I decide about the future procedure based on the reviewer comments and your answers to these, including your cover letter, and then potentially invite you to revise and resubmit your manuscript. I therefore have not considered your revised manuscript at this stage. However, given that you have published your revised manuscript as author comment, we are not in the position to remove this anymore from the web-page.

I recommend that your manuscript undergoes major revisions as outlined by the reviewer comments. I would emphasise in particularly the need to strengthen the discussion section in light of the limitations of the experimental set-up.

Best wishes,
Sönke Zaehle

Dear author,

many thanks for your revised version and my apologies for the delay in providing an assessment of this version due to health reasons.

While I find that your manuscript much improved, there remain several areas of concern, requiring a new revision to be acceptable for Biogeosciences. My main concerns are

- you added fish-eye observations, but they need to be mentioned as a separate section in the results (including a figure or a table), and not buried in the discussion.

- your text reads as if there had been largely no or a negative response, but this does not reflect Figure 2 panel c which suggests that the growth rate of large trees has almost doubled. Maybe this is not statistically significant, because of the large variation, but this is too large a difference to be ignored. This is one of the points where the limitiations of the study design need to be more carefully addressed, as the small plot size and low replication reduce the statistical power of the experiment. It is important to mention the limitation of the design not only in the Methods section, but also in case it does affect the interpretation of the Results.

- Figure S1 provides good information in addition on what is going on at the sites, and should not be buried in the SI. Please include this, and a paragraph outlining the implications of this data into the main manuscript.

- As reviewer#1 already commented, please remove the reference to the unpublished material of the P fertilisation experiment (e.g. L289-298), because without proper explanation of the experimenet design, results and their uncertainty, these data appear unsubstantiated.

- As reviewer #1 already commented, the environmental relevance of the magnitude of the fertilisation is questionable. Please down-tone any reference to N deposition in the introduction (e.g. first paragraph), as your experiment is not designed to address the likely response of N deposition on these ecosystems

Best wishes,
Sönke Zaehle

Dear authors,
many thanks for your revisions. Unfortunately, I need to ask you again for some - this time minor - corrections, before the manuscript can become acceptable. Please see my comments below.
Best wishes,
Sönke Zaehle

- the title focusses the comparison on growth form, whereas in reality, your results mainly give evidence for the differential response between over and understorey, with the latter including different growth forms. This needs to be adequately reflected in title and abstract.

- In the methods, you have removed the information on site N deposition, which is however, relevant. I would present this information next to precipitation and temperature estimates earlier in Section 2.1

- New Section 3.1: I think this section needs to be accompagnied by a statement as to whether it is possible that the fish eye measurements did not provide evidence for changes in total forest cover, while there still may be a shift between the contribution of overstorey and understorey trees to the total forest cover (partly therefore explaining the difference between the response across DBH?

- Given that the pH at the end of the treatment is 4.1, whereas in Section 2.1 it was introduced as 4.6, this needs more explicit mentioning in Section 3.1 and the Discussion on other limiting factors. This is a change of 0.5 pH within 3.4 years, which in my view cannot be described as "mild".